Free Will and Determinism
Michael Norwitz examines the current state of play in this long-running debate, by comparing the views of Dennett and van Inwagen.
Since the ancient Greeks, one of the most provocative and oft-discussed questions in philosophy has been whether we have free will in determining the course of our actions, or whether our actions are determined by forces beyond our control. Before the advent of secular thought, those forces might have been identified as the whims of the gods, though the tradition of naturalism in Western thought goes back at least as far as the Milesian School of Greek Philosophy, in the 6th century B.C. In more recent times as the cognitive sciences have developed, it has seemed increasingly likely that our brains work along deterministic lines (or, if quantum effects are non-negligible, at the very least along mechanical lines). So a new debate has arisen: are the concepts of determinism (or naturalism or mechanism) when applied to the brain sciences logically compatible with free will? So some of the attention has shifted from the debate between the “determinists” and the “anti-determinists”, to that between the “compatibilists” and the “anticompatibilists”.
Two declared opponents in this debate are Peter van Inwagen (author of An Essay on Free Will, Oxford University Press, 1983) and Daniel C. Dennett (author of several books including Elbow Room, MIT Press, 1984, which I will be referencing here). Each argues for his conclusion from premises he regards as antecedently plausible, with van Inwagen taking the anti-compatibilist line and Dennett the compatibilist. As van Inwagen is the more precise arguer of the two, I will use his work as the starting point for this discussion. Like Dennett, whose book is subtitled “The Varieties of Free Will Worth Wanting”, he is arguing that we do have free will.Where they differ is on the nature of its relationship to determinism. Van Inwagen presents three premises in his main argument : that free will is in fact incompatible with determinism, that moral responsibility is incompatible with determinism, and that (since we have moral responsibility) determinism is false. Hence, he concludes, we have free will.
The argument for the first premise runs as follows [p.56]: “If determinism is true, then our acts are the consequences of the laws of nature and events in the remote past. But it is not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore the consequences of these things (including our present acts) are not up to us.”
The argument for the second premise [p. 181]: “If (i) no one is morally responsible for having failed to perform any act, and (ii) no one is morally responsible for any event, and (iii) no one is morally responsible for any state of affairs, then there is no such thing as moral responsibility.”
For the third premise van Inwagen does not present a concise summary of his line of argument. He takes it as being self-evident that we have moral responsibility, as we do, after all, continue to hold people morally responsible for their actions.
Dennett would not fault the validity of van Inwagen’s main argument; he does argue with the truth of its premises however. His approach is to reformulate the concepts of “up to us” (in the sense of the argument for the first premise) and “responsibility”. Before I expand on that, however, I want to discuss what I think is the difference in the philosophers’ starting points that causes the divergence of opinion.
Descartes viewed the mind as a pure ego: a permanent, spiritual substance untouched by physical processes. It could be influenced by them through the senses but there was no other manner in which it was influenced by the mechanistic events going on outside in the world. It could influence those events indirectly however through the manipulation of its host body (via the pineal gland).
As modern science advanced in its understanding of the way the brain works, this image of the mind was undermined. It began to look more and more as if the mind is a purely physical entity, as if there is no “person” or “pure ego” outside the realm of physical causation. Some philosophers (like the Churchlands) now go so far as to say that the mind does not exist at all.
In the face of this, the philosopher of metaphysics has two options: retrenchment and retreat.
Dennett’s strategy of retrenchment is to build a second line of defence for the concept of free will, by reformulating the concept so that it is not in conflict with current theories in the brain sciences. There is a sacrifice in that he loses track of our ordinary, common-sense views of what mind and free will are. Dennett claims he is doing ordinary language philosophy but I suspect he has been an academic so long he has forgotten what “ordinary people” are concerned with.
Van Inwagen’s strategy of retreat is to dismiss current trends in science and maintain belief in “agent causation”, that is, the view that people can cause things to happen in the world outside of the normal, mechanistic, physical causation. He complains that many philosophers are overawed by current science and make exaggerated assumptions about the degree to which it will eventually be able to explain how the brain (and the mind) works. However, for various reasons, chief among them being the empirical success of quantum physics, it is highly unlikely that such a complete explanation will ever come about. Heisenburg’s Uncertainty Principle, if it can be applied to the brain, would mean that even if we knew everything about the physical state of a brain at a given instant, we still could not predict its state in the next instant with absolute accuracy. This would imply that the brain was not deterministic in the strictest sense of the term. Nevertheless, as van Inwagen correctly points out, even were determinism false there would still be no guarantee that we have free will. First, if our hopes turned on quantum effects being able to affect brain chemistry, it is still conceivable that they might turn out to be too small to be significant. Second, even if they did have an effect which was non-negligible we could still turn out to be strictly mechanical, and that does not seem to be the type of free will that van Inwagen wants, if he wants a “person” making responsible decisions free from causal restraints (at least physical causal restraints, as he accepts psychological causation).
Ultimately, van Inwagen states that we know we have free will because free will is entailed by moral responsibility, and we know that people are morally responsible for their actions. The rationale for this entailment is van Inwagen’s conception of moral responsibility [p.162]: “a person is morally responsible for what he has done only if he could have done otherwise” (his final version of moral responsibility is more baroque to deal with various styles of counterexamples, but this much simpler version is sufficient for our purposes).
Dennett claims there are cases of responsible action when one could not have done otherwise. That is the purpose of a moral education, to make one incapable of, say, torturing an innocent person in exchange for a thousand pounds. We may have been trained since birth to consider such an offer unacceptable, yet most of us would not claim when we rejected the offer we were not doing so freely. Dennett asks, what is it we want to know of a person when we wonder, could he have done otherwise in a particular situation? Are we asking, given the exact brain states he had and the exact state of the universe as it was at the time of the act, could the person have done otherwise? Dennett rejects this formulation of the question as unanswerable, and even if answerable as unhelpful in determining responsibility. Unanswerable because it is impossible for us to duplicate a model of such complexity; unhelpful because even could we by some stretch of the imagination lay out such a model, we will never naturally find ourselves in such a state – even were the external condition the same the cognitive conditions would not be (at best we might experience some sense of déja vu). So we are left with the problem of how to interpret the question so that it does illuminate [p.142]:
We ask [the question] because something has happened that we wish to interpret … we want to know what conclusions to draw from it about the future. Does it tell us anything about the agent’s character, for instance? Does it suggest a criticism of the agent that might, if presented properly, lead the agent to improve his ways in some regard? Can we learn from this incident that this is or is not an agent who can be trusted to behave similarly on similar occasions in the future? If one held his character constant, but changed the circumstances in minor – or even major – ways, would he almost always do the same lamentable sort of thing? Was what we have just observed a “fluke”, or was it a manifestation of a “robust”trend – a trend that persists, or is constant, over an increasingly wide variety of conditions?
Thus, Dennett argues, we would still hold people morally responsible whether we accepted van Inwagen’s concept of free will or not, because the considerations we have in mind when we ask whether someone “could have done otherwise” are irrelevant to issues of free will and determinism.
I doubt van Inwagen would be satisfied with Dennett’s approach. Despite its ingenuity it comes off like a verbal trick; it “solves the problem“ but at the cost of not really approaching what we worry about when we worry whether we have free will, or responsibility. Of course, Dennett would respond that these worries are bugbears.
That, I think, is a manifestation of the fundamental disagreement. Resolving this disagreement would help resolve the issue between them about free will, but I have my doubts over whether any such resolution is possible. Their disagreement is based on a fundamental judgment each of the two has made about how philosophy should respond to the other disciplines around it.
I agree with van Inwagen’s observation that, given the current state of science, it is premature to claim that determinism (neurologically if not cosmologically) is true; however,it is certainly premature to claim that it is false as well. I see no reason to be convinced by van Inwagen’s arguments unless he is able to give some vague picture of how he thinks agent causation might physically work. I don’t expect it to be exact, but he ought to at least be able to tell a convincing story. The compatibilists can tell a very interesting story, though we might not care so much for their conclusions. Without some kind of workable story, so far as I can tell, van Inwagen is tacitly accepting Cartesian egos as the source of our free will. He is well aware of this shortcoming but is not overly bothered by it. I think that falling back on the Cartesian model and trying to operate outside the realm of empirical science is not a sacrifice worth making. Dennett’s recommendations are worth taking seriously, despite his apparent lack of awareness of the sacrifice he makes in abandoning our ordinary concept of free will – I think this is a sacrifice worth making.
© Michael Norwitz 1991
Free Will and Determinism
Oh Thou, who didst with Pitfall and with Gin
Beset the Road I was to wander in,
Thou will not with Predestination round
Enmesh me, and impute my Fall to Sin?
Rubáiyát of Omar Khayyám
If it is possible even in theory to predict all that will happen in the future, does this mean that we are not free? Could we look at a child and by considering all the forces and influences which we knew would act upon him over the years, predict accurately that the child would grow up to be a serial killer? And if, in principle, we could, does this mean that the serial-killer-to-be is not responsible for his actions?
The debate about free will and determinism has been going on for centuries. It affects all our ideas about morality and human actions. This issue of Philosophy Now contains two articles on the topic. The first, by Michael Norwitz, sets the scene by examining the ideas of two current participants in the debate. The article which follows is an original contribution to the debate by Professor Antony Flew.